Dinamik ağların eniyilenmesi

Many systems in engineering, sciences and management can be represented as networks. The network structures can be encountered in many real world applications. Although networks are generally considered as a stationary snapshot of a system, most of the networks exhibit dynamic behavior. In this thesis, the main objective is the development of mathematical models for dynamic networks and the development of solution algorithms to optimize the corresponding dynamic networks. Two different dynamic networks are considered in this thesis: Supply chain networks and metabolic networks. In the supply chain networks part, the concept of price elasticity of demand is superimposed into the dynamic supply chain network. The resulting nonlinear programming problem is solved using two approaches. In the first approach, the problem is formulized as a convex quadratic optimization problem and solved as an NLP problem. In the second approach, piecewise linear approximation is used to solve the system. In addition, dynamics of the wine fermentation is modeled using the metabolic network of the Saccharomyces Cerevesiae. The feedback relation between cell interior and exterior is integrated that gives rise to an optimization problem with Differential and Algebraic constraints. The integrated model is solved by hybrid system approach to estimate the parameters in the fermentation dynamics of wine.